GPT 4.1 nano is a small and intelligent model from OpenAI for fast, lightweight tasks.

Anthropic's fastest model that can execute lightweight actions, with industry-leading speed.

A 32-billion-parameter model from Alibaba's Qwen team, offering advanced reasoning capabilities comparable to larger models, excelling in math, coding, and problem-solving tasks.

Ideal for intermediate tasks that require moderate reasoning.

Google's fastest, most cost-efficient multimodal model with great performance for high-frequency tasks

Meta's 70-billion-parameter multilingual model optimized for dialogue, demonstrating enhanced performance in reasoning, mathematics, and instruction-following tasks.

Anthropic's latest model balancing performance and speed, suitable for complex analysis, multi-step reasoning, and advanced math and coding tasks.

Power your assistants with the latest models from OpenAI, OpenAI’s most advanced system, producing safer and more useful responses. GPT-4 surpasses ChatGPT in its advanced reasoning capabilities.

An advanced language model designed for tasks requiring moderate reasoning, with improved performance in text generation, summarization, and question-answering.

An open-source reasoning model rivaling leading counterparts, excelling in mathematics, coding, and complex reasoning tasks, offering a cost-efficient solution for high-level analytical needs.

Meta developed and released the Meta Llama 3.1 family of large language models (LLMs), a collection of pretrained and instruction tuned generative text models in 8 and 70B sizes.

Google's best performing multimodal model with features for a wide variety of reasoning tasks

Mistral's flagship model that's ideal for complex tasks that require large reasoning capabilities or are highly specialized